Dr. Prausnitz has testified in numerous Duragesic fentanyl patch lawsuits, including this case, which ended in a $16 million dollar verdict for the plaintiff.

Lots of good information in this deposition about the defects in the Duragesic patch design, and the dangers of fentanyl and fentanyl gel.

THE VIDEOGRAPHER: On the record.

MARK R. PRAUSNITZ, Ph.D., having been first duly sworn, was examined and testified as follows:



Q. Good morning, sir.

A. Good morning.

Q. Could you please state your name for the record.

A. My name is Mark Prausnitz.

Q. Dr. Prausnitz, this is the second time we've met. I was here late last year, and we deposed you in a case called Burchill. We talked briefly off the record. I've gone through some of the materials you brought with you. We've talked about what you may refer to; and, as I mentioned before we went on the record, I'm not going to spend a whole lot of time going over material that's already been covered with you, but we may touch on some of that material. Okay?

A. Okay.

Q. Okay.

A. In terms of matters of more scientific content, calculations, data, significance of data, that's where there could be some, some differences of interpretation.

Q. Okay. Now — and I don't mean to cut you off if you had more to add on to that.

A. Well, so one topic where I did not agree had to do with some estimates, predictions, of what would happen if Duragesic gel escaped the patch and went onto the skin.

Q. And it would certainly be great if you agreed with us. Perhaps we wouldn't be here today. But I would assume that there would be some disagreement on that.

What specifically — let me back up and ask it this way: Do you disagree with his ultimate conclusion as to what would happen if gel got on the skin?

A. I think —

MR. ORR: Objection, form.

THE WITNESS: I think we can agree that if gel got on the skin fentanyl would enter the body for a period of time. At some period of time later, the gel, if exposed to the air, would evaporate. I think we disagree in terms of time scales of evaporation, degree of absorption.

Q. (By Mr. Naeem) With respect to the historical information Mr. Gale relied on to support his opinions regarding the time scale of evaporation and degree of absorption, do you have any information to suggest that what he was relying on was inaccurate?

A. Regarding evaporation, I think what he relied primarily on was a particular study where gel was put into an oven and evaporation rates were measured. And there was some discussion in the deposition about the validity of those experiments, and he admitted that it was an imperfect experimental design. He nonetheless concluded that it was appropriate for him to rely on it. My feeling is that the flaws were — the flaws in the experimental design were significant and made a serious overprediction of the rate of evaporation.

Q. Okay. So that's a criticism of the methodology of something he relies on. But my question was: Was there anything that he relied on to support his opinions, any of the historical

[Note: Pages 14-76 missing in original document]


A. That's right.

Q. All right. So in your estimation, am I paraphrasing your representation correctly that it would be somewhere between the 20 percent line and the 30 percent line?

A. That's right. And fortunately those lines are very close.

Q. Okay. And can you give me an estimate as to what you think that number would be?

A. That, if we compare the 10 percent line to the 20 and 30 percent lines, it looks like another third or 50 percent type of increase.

Q. Okay.

A. So I would expect, if you used both what I believe is the correct temperature and the correct ethanol concentration, you wouldn't get a 30 percent increase, you would get more like a doubling.

Q. Okay. And, I'm sorry, so it would double when you factor in the temperature with the use of 23 percent ethanol. What was — you said a third, not factoring for the temperature?

A. Roughly a little over a third, it looks to me.

Q. Okay.

[Note: Pages 78-97 missing in original document]

the effects of evaporation, and I do want to turn to your opinions that deal with evaporation. And in an attempt to perhaps at least blow through some of the preliminary stuff, what I want to do is is turn back to your last deposition and just read a couple statements that you made and see whether you still agree or disagree with those, okay.

First of all, I asked you whether you agree with me that ethanol and water in the Duragesic gel are subject to evaporation; and you said yes. Do you still agree with that?

A. Yes.

Q. Okay. I asked you whether — and I'll just read directly what my question was and your answer. So once evaporation is complete, are you saying the fentanyl actually precipitates on the skin? Your answer: That's my expectation, yes. Do you still agree with that?

A. If there is no liquid on the skin, then the fentanyl has nothing to dissolve in; and, therefore, it would be a solid precipitant.

Q. Okay. And so my question was: Evaporation is complete, precipitation occurs? You still agree with your answer, which was?

A. Well, I guess I'd add to that to say that there could be a source of liquid to dissolve fentanyl other than the original gel. There could be oil from the skin, there could be sweat, and perhaps another source as well. So discounting any other source of liquid, then once the liquid from the gel is gone, there's no more liquid there.

Q. Okay. I asked you: Once evaporation is complete, the only fentanyl that can be absorbed into the blood is what has already crossed into the skin, that is — that is, is in that depot in the skin. And your answer was yes. Do you still agree with that?

A. Can you read the first few words of that sentence or question again?

Q. Once evaporation is complete, the only fentanyl that can be absorbed into the blood is what has already crossed the skin, that is, in that depot in the skin? And you said yes?

A. I would still agree, again, with the caveat that it assumes that there is no other liquid present that could dissolve fentanyl.

Q. Okay. In the same manner you described the oil on the skin —

A. That's right.

Q. — or the moisture? Okay. We were talking about the Acrux and the Sebel studies, and I asked you: What they don't do is measure how long it takes ethanol and the water to evaporate from the solution that was dried onto the skin. Your answer was: They are measuring the fate of drug. They are not measuring the fate of solvent. Do you still agree with that?

A. I'm sure they have measured evaporation rates. I'm sure they know something about that. In the documents that I had, that wasn't quantified.

Q. And then I asked you: So out of the three things you mentioned, the Acrux data, the Sebel article, and the ALZA study, the only one that actually quantifies or attempts to quantify the rate of evaporation is the ALZA study? And you said that's right. Do you still agree with that?

A. Among — so the ALZA study I assume being the study measuring evaporation rates in an oven?

Q. Right.

A. Among those three studies, that is the only one that makes measurements of evaporation.

Q. Okay. And have you reviewed or cited to any other literature in preparation for your opinions in this case which attempt to quantify evaporation, the rate of evaporation of ethanol in water from a solution similar to Duragesic?

[Note: Pages 101-154 missing in original document]


A. I can't give you an exact number, quantitative number, for what the evaporation rate of Duragesic gel is because it is a function of conditions. So there isn't a number. It's a mathematical function. I do expect that in situation that is relevant to the DiCosolo case and in general to fentanyl gel that's on the skin that there has been an overestimate of the rate of evaporation that is at least manyfold, maybe tenfold.

Q. But based on what you know, you can't quantify that? I just want to be clear. I'm not — again, I'm not being critical; but I need to know if you're going to go into a courtroom and quantify it and say based on my calculations that number would be not 2.2 but it would be 22.2 — maybe that would be the wrong way. It would be 0.2.

A. I am not going to testify that I know quantitatively the evaporation rate of gel from Ms. DiCosolo's skin. I will because of lack of data have to talk in more general terms.

Q. Okay. And then one more question on that. If you designed this study to control these various factors we talked about, is it your estimation that the evaporation time for the various scenarios presented in the report would go from minutes to hours?

A. The report concluded minutes, as I recall. If you are to have no — no airflow, no forced airflow, were to have a humidity that was representative of an ambient environment, to have the temperature be correct and the heat transfer be correct for the skin, I guess I'll go back to what I said before, which is I think that would — that would decrease the rate of evaporation at least manyfold, maybe tenfold. So I think it's quite possible that it could be hours, especially if you put the added possibility of there being an overlay or if there is clothing or some other kind of occlusion that could slow things down even further.

MR. NAEEM: Okay. I'm at that point where I would be getting into the supersaturation, so we can go off the record.

THE VIDEOGRAPHER: Off the record.

(Off the record.)

MR. NAEEM: The parties have agreed to adjourn Dr. Prausnitz' deposition until Thursday, February 28th, at 5:30 p.m. On behalf of defendants, this is Tariq Naeem, I have agreed to limit my deposition to topics not covered, which

[Note: Pages 157-159 missing in original document]


February 28, 2008

6:10 p.m.

1776 .Peachtree Street

Suite 230-S

Atlanta, Georgia

Linda C. Ruggeri, CCR-A-261


On behalf of the Plaintiff:


Heygood, Orr, Reyes, Pearson & Bartolomei

2331 W. Northwest Highway

Second Floor

Dallas, Texas 75220



Angwin Law Firm

The Black Diamond Building

2229 First Ave N

Birmingham, Alabama 35203-4203


On behalf of the Defendants:


Tucker, Ellis & West

1150 Huntington Building

925 Euclid Avenue

Cleveland, Ohio 44115-1475



Mr. Allen Miegel

(Reporter disclosure made pursuant to Article 8.B. of the Rules and Regulations of the Board of Court Reporting of the Judicial Council of Georgia.)

THE VIDEOGRAPHER: On the record.

MARK R. PRAUSNITZ, Ph.D., having been previously duly sworn, was examined and testified further as follows:



Q. Dr. Prausnitz, good evening. We're back on the record. You understand that you're still under oath with respect to the testimony you're giving in this case?

A. Yes, I do.

Q. Okay. When we broke for the evening last time I was down in Atlanta, we had been talking about your opinions regarding the DiCosolo case; and specifically we had finished talking about the effect of evaporation on your opinions in this case. What I want to move on to now is some of the opinions you've given or some of the comments you've made about something called supersaturation. Okay?

A. Okay.

Q. And specifically on Page 7 of your report, you state that the — I'm sorry. You state that during solvent evaporation the fentanyl concentration can dramatically increase due to supersaturation of the drug in the ethanol formulation. Did you get that, Doctor?

A. I see where you are. It's under the underlined text.

Q. Yes, it is, underlined middle of the page.

A. Yes.

Q. Is that still your opinion in this case?

A. Yes, it is.

Q. Okay. Are you saying — you say that fentanyl concentration can dramatically increase. Are you saying it's possible or are you saying it's probable?

A. I'm saying — I'm certainly saying that it is possible. I'm saying that it would also provide a reasonable explanation for the absorption that we see not just in this case but absorption that is seen in other studies that we've discussed.

Q. Okay. And, now, you have a number of citations within the body of your report there about supersaturation; but you just said it would support not just what we've seen in this case but what you've seen in other sources. I'm not sure I understand what

[Note: Pages 164-175 missing in original document]

in your report, it's the Coldman article.

A. Yes, I see it.

Q. Okay. Can you explain to me how this article informed your opinion?

A. This informed my opinion in a similar way to the Megrab article that we just discussed. It is another article that is representative of the larger literature on supersaturation as a means for delivery.

Q. Okay. Are you aware of whether there was any measurement in this article as to the length of time that supersaturation would occur from application of this particular solution to the skin?

A. I don't believe that time scale was measured.

Q. Okay. The Megrab or the Coldman articles, do they deal with fentanyl, either one of them?

A. Neither of them is about fentanyl per se; but both of them are about molecules with similar properties — I should say physical chemical properties, not pharmacological properties — and, therefore, the results from these studies are directly representative or related to what you would expect to happen with fentanyl.

Q. Okay. And Megrab, of course, evaporation wasn't — supersaturation as a result of evaporation

[Note: Pages 177-185 missing in original document]

specific scenario, is that, ones that I know of, the measurements have not been made as to the degree of supersaturation that occurred.

Q. Doctor, do you have an opinion to a reasonable degree of scientific certainty regarding whether the fentanyl solution in Duragesic gel on the skin would become supersaturated as that solution evaporated?

A. Could you say that again?

Q. Yeah. Do you have an opinion to a reasonable degree of scientific certainty regarding whether the fentanyl solution in Duragesic gel on the skin would become supersaturated as the solution evaporated?

A. I can say with certainty that that system is capable of supersaturation. I can't say with certainty that, for example, in the case that we're discussing that supersaturation occurred or to what degree it occurred.

Q. Okay. Doctor, do you have an opinion to a reasonable degree of scientific certainty regarding whether the fentanyl solution from Duragesic gel on the skin, how long that solution would remain supersaturated?

A. I don't have the data that I need in order

[Note: Pages 187-225 missing in original document]

similarly high level — I don't know that it's, you know, to three significant figures the same, but some similarly high level that existed previously.

Q. And what is that you're referring to?

A. Again, it's the medical examiner and it's also there were experts, other experts, that Mr. Angwin has retained in this case that provided similar opinions.

Q. Okay. And you reviewed those reports?

A. I have reviewed those reports.

Q. Okay. Are you going to be produced — I'm sorry, providing any opinions in this case as to what caused the 28.2 level measured 24 hours' postmortem in Janice DiCosolo's blood?

A. I will be providing opinions as to what caused a highly elevated level that is — again, I'll use that phrase — in the neighborhood, is some high level such as 28.2. But I'm not going.to try to make an exact quantitative calculation where I say, well, this evaporation effect and this temperature effect and this area effect and they all add up exactly to 28.2 nanograms as a resulting concentration.

Q. Okay. And what is the opinion that you're going to be giving as to the cause of Janice DiCosolo's fentanyl blood level after death?

[Note: Pages 227-230 missing in original document]

would have happened. The number of patches were accounted for as I understand it, and there was no indication that such an incorrect use occurred.

Q. Is it possible to get to a 28.2 level from the application of heat to a Duragesic patch?

MR. ANGWIN: Object to the form. And, Tariq, for your questions, is he supposed to assume any particular strength of patch?

Q. (By Mr. Naeem) Let's assume a 75-microgram patch similar to the one being worn by Janice DiCosolo.

MR. ANGWIN: I not trying to interfere. I just want to make sure the questions are clear.

MR. NAEEM: That's fine; that's fine.

Q. (By Mr. Naeem) Do you want me to rephrase the question?

A. No, I think I understand. So if one applies heat to a patch, that can increase the rate of delivery. And there have been some studies showing increases. But to get to a 28.2, you need an extremely large increase. The 28.2 is about 16 times larger than the expected average level. And I don't believe that those heat studies have shown that kind of an increase.

Q. What is, Doctor, your definition of the

[Note: Pages 232-235 missing in original document]

Q. Okay. Any evidence — do you have any evidence that Janice DiCosolo had greater-than-average skin permeability?

A. There is a 50 percent chance that she did.

Q. Do you have any evidence that she did?

A. I do not know what her skin permeability is.

Q. Any evidence that Janice DiCosolo had damaged skin under or around the area she placed her patches?

A. There is no evidence for that other than the fact that a lot of fentanyl entered her body more quickly than it should have.

Q. Okay. We've talked about a lot of facts regarding Janice DiCosolo that supports your — that you're testifying supports your opinion regarding her fentanyl blood levels. Are there any other facts that you've reviewed and that you are relying on to support your opinion that the 28.2-nanogram-per-milliliter level was caused by a defective Duragesic patch or patches?

A. The information that I have relied upon is in my expert report and has been discussed over the course of this deposition.

Q. Okay. I want to take you through your ??
Expert Declaration from Mark R. Prausnitz, Ph.D.

I am a Professor of Chemical and Biomedical Engineering at the Georgia Institute of Technology and at Emory University Medical School. I also serve as the Director of the Center for Drug Design, Development and Delivery at the Georgia Institute of Technology. I received a B.S. degree in chemical engineering from Stanford University and a Ph.D. degree in chemical engineering from the Massachusetts Institute of Technology (MIT). I have carried out research in the field of transdermal drug delivery since 1988 and currently have an active laboratory that continues to work in this field. My full curriculum vitae is included at the end of this report.

I have been asked to assess the degree to which fentanyl can be absorbed across the skin after leaking from a defective Duragesic patch. To make this assessment, I have reviewed many studies from the scientific literature, including those cited in this report. There is significant evidence demonstrating that large amounts of fentanyl can be absorbed across the skin during an uncontrolled dermal exposure to an ethanol gel formulation of fentanyl, such as that leaking out from a defective Duragesic patch. This leakage could result in absorption of a lethal dose of fentanyl.
I. Overall conclusions

I have assessed the degree to which fentanyl can be absorbed across the skin after leaking from a defective Duragesic patch. My conclusions are as follows.

(1) Fentanyl absorption across the skin is increased in the absence of a rate-controlling membrane.

(2) Skin permeability greater than the average value due to normal statistical variation or the possibility of damaged skin would result in still greater increases in fentanyl delivery, which would have been largely prevented by the safety afforded an intact Duragesic patch.

(3) A lethal fentanyl dose may be achieved individually or by a combination of the effects of (i) loss of the rate-controlling membrane, (ii) increased skin permeability, (iii) increased skin area contacting fenantyl gel and (iv) supersaturation of the fentanyl gel formulation.

(4) While ethanol enhances skin permeability to fentanyl, significant fentanyl delivery can be achieved across the skin in the absence of ethanol.

(5) Application of fentanyl to the skin using a formulation solvent that rapidly evaporates can lead to extensive fentanyl absorption over a period extending for many hours, as shown in a number of studies involving human subjects. The rabbit study conducted by Alza and discussed below failed to demonstrate this behavior probably due to the limitations of using rabbit skin.

(6) Quantitative analysis of the effects of various parameters on skin permeability demonstrates the ability to achieve serum fentanyl concentrations known to be fatal due to leakage of fentanyl gel from a Duragesic patch.

(7) A properly functioning Duragesic-75 patch should produce a fentanyl blood level between 1.0 and 2.4 ng/mL. A fentanyl blood level of 28.2 ng/mL should not be achieved using a properly functioning patch. A patch that produces this high level of fentanyl is most probably defective.

(8) The reservoir design of the Duragesic patch is dangerous, because drug can leak if the patch malfunctions, which can cause a drug overdose. Similar fentanyl patches that utilize the matrix design cannot leak and therefore do not pose this risk. Such safer designs have been under development at Alza for many years, represented a feasible commercial alternative at the time of the decedent's death, and would have eliminated the risk of catastrophic patch failure due to leakage.

(9) In the present case, I conclude that the high level of fentanyl found in the blood of Janice DiCosolo was the result of a leak or other defect in one of the Duragesic patches she was wearing prior to her death. That is because the fentanyl concentration in her blood was 28.2 ng/mL, which is more than 16 times larger than the average maximum concentration from a properly functioning Duragesic-75 patch, i.e., 1.7 ng/mL. Moreover, there is no evidence of abuse, misuse or altered fentanyl clearance by the decedent.

A detailed analysis supporting these conclusions follows.
II. Detailed analysis of fentanyl absorption from a defective Duragesic patch

My analysis starts with an assessment of the intended performance of a non-defective Duragesic patch. According to the manufacturer's prescribing information, the Duragesic-75 patch is supposed to achieve a peak serum fentanyl concentration between 1.0 and 2.4 ng/mL during its intended 72-hr application time.[FN1] The Medical Officer Review carried out by the FDA found that just 3% of Duragesic-75 patches produced fentanyl concentrations greater than 3.0 ng/mL and that none of them exceeded 5.0 ng/mL, which is referred to as “dose dumping.”[FN2] A fentanyl blood level of 28.2 ng/mL should never be achieved. For comparison, lethal fentanyl concentrations have been found to range from 3.0 to 28 ng/mL, with an average value of 8.3 mg/mL.[FN3] This indicates that even a few-fold increase in fentanyl delivery rate above that achieved by a non-defective Duragesic patch can be fatal and an order-of-magnitude increase in fentanyl delivery rate would almost certainly be fatal.

    FN1. Physicians' Desk Reference (2006) Thompson PDR, Montvale, NJ

    FN2. Medical Officer Review, NDA#: 19,813, ITS Fentanyl, Volume 2 – Pharmacokinetics & Pharmacodynamics, 1990.

    FN3. RC Baselt, Disposition of Toxic Drugs and Chemicals in Man, 7th edition, Biomedical Publications, Foster City, CA

These dangers are explicitly and repeatedly recognized in the package insert for Duragesic prepared by Alza and approved by the FDA, as summarized below.[FN4]

    FN4. Physicians' Desk Reference (2006) Thompson PDR, Montvale, NJ

• The black box warning states “Using damaged or cut DURAGESIC(R) patches can lead to the rapid release of the contents of the DURAGESIC?? patch and absorption of a potentially fatal dose of fentanyl.”

• The Description section repeats “If the DURAGESIC(R) system is cut or damaged, controlled drug delivery will not be possible, which can lead to the rapid release and absorption of a potentially fatal dose of fentanyl.”

• The Warnings section repeats “Using damaged or cut DURAGESIC(R) patches can lead to the rapid release of the contents of DURAGESIC(R) patch and absorption of a potentially fatal dose of fentanyl.”

• The Warnings section further states “Death and other serious medical problems have occurred when people were accidentally exposed to DURAGESIC(R). Example of accidental exposure include … possible accidental exposure of a caregiver's skin to the medication in the patch while the caregiver was applying or removing the patch.” This indicates that even very brief exposure can be fatal.

• The Precautions section repeats “The DURAGESIC(R) patch should not be used if the seal is broken, or if it is altered, cut, or damaged in any way prior to application. This could lead to the rapid release of the contents of the DURAGESIC(R) patch and absorption of a potentially fatal dose of fentanyl.”

• The Drug Abuse and Addiction section repeats “Using cut or damaged DURAGESIC(R) patches or its contents can lead to the rapid release and absorption of a potentially fatal dose of fentanyl.”

• The Dosage and Administration section repeats “Using damaged DURAGESIC(R) patches can lead to the rapid release and absorption of a ??otentially fatal dose of fentanyl. In addition, exposure to the contents of a DURAGESIC patch can lead to potentially fatal respiratory depression.”

• The Dosage and Administration section also states “If the gel from the drug reservoir accidentally contacts the skin of the patient or caregiver, the skin should be washing with copious amounts of water.”

• The How Supplied section repeats “If the gel from the drug reservoir accidentally contacts the skin, the area should be washed with copious amounts of water.”

• The How Supplied sections also repeats “If the DURAGESIC(R) system is cut or damaged, controlled drug delivery will not be possible, which can lead to the rapid release and absorption of a potentially fatal dose of fentanyl.”

As shown by these examples, the manufacturer has felt it necessary to repeatedly provide warnings about the dangers of using a defective patch in ten different locations in the product insert and to specifically mention in eight of them that use of a damaged patch can be fatal. This clearly shows that the manufacturers of DURAGESIC(R) are seriously concerned that uncontrolled fentanyl delivery from a defective patch is extremely dangerous and possibly fatal.

Given these dangers, transdermal patches, such as Duragesic, are designed to have accurate control over the rate of drug delivery into the skin. For example, a non-defective Duragesic-75 patch delivers fentanyl across the skin at a rate of 75 mcg/hr from a patch reservoir containing fentanyl at a concentration up to 25 mg/mL.[FN5] However, defense counsel informed me that only 9.8% of the fentanyl is actually in solution.[FN6] If we accept this information, even though defense counsel did not provide any documentation to support it, we get an effective fentanyl concentration of 2.45 mg/mL. A study from the University of Michigan measured the rate of fentanyl transport across human cadaver skin and found it to be 5.6 x 10-3 cm/hr.[FN7] For a fentanyl solution of 2.45 mg/mL and a skin contact area of 30 cm2 (which is the skin contact area for Duragesic-75), this skin permeability corresponds to a delivery rate of 412 mcg/hr. This demonstrates that the rate-controlling properties of a non-defective patch can be significant and a loss of those properties can dramatically increase fentanyl delivery rate. which could be fatal.

    FN5. Physicians' Desk Reference (2006) Thompson PDR, Montvale, NJ

    FN6. Deposition of Mark R. Prausnitz, Jason Burchill, et al. v. Johnson & Johnson, et al., September 18, 2007.

    FN7. SD Roy and GL Flynn, Pharmaceutical Research (1990) 7:842-847

It is worth noting that in the above analysis and the analysis that follows in the rest of this report, almost all of the studies cited have used human skin, which is the best model to predict behavior in humans. Some of the studies have used human cadaver skin, which differs from living human skin in a number of ways, but is nonetheless widely accepted in the research community as an excellent model to determine skin permeability that correlates well with skin permeability in human patients. These studies also all come from leading universities or respected companies (including Alza and Janssen) and have been published in peer-reviewed journals, presented at international scientific conferences, or disclosed in issued US patents. I therefore believe that the data used in this analysis are of high quality and reliability.
II.a. Variability of skin permeability to fentanyl

The above calculation has assumed average skin permeability values, but skin permeability has considerable inter-subject and intra-subject variability. To address this variability, the Duragesic patch was designed to include a rate-controlling membrane to reduce the impact of variable skin permeability on the rate of drug delivery. In a journal article written about the fentanyl patch by employees of Alza (the company that designed the Duragesic patch) and collaborators at Stanford University, it states that the “permeability of the stratum corneum varies widely” and that use of a rate-controlling membrane “allows the overall control of drug absorption to reside in the design of the transdermal system and not be dependent upon the widely variable permeability of the stratum corneum”.[FN8] The stratum corneum is the outer layer of skin that controls the rate of dermal absorption of fentanyl in the absence of a rate-controlling membrane.

    FN8. JR Varvel et al., Anesthesiology (1989) 70:928-934

In the case of a defective Duragesic patch, the fentanyl formulation can come out of the patch housing and make direct contact with the skin. In this case, the rate controlling membrane is no longer present and the widely variable permeability of the stratum corneum becomes an important determinant of the amount of fentanyl delivered. In this scenario, those people with highly permeable skin are especially at risk.

Simply removing the rate controlling membrane should at least double the fentanyl delivery rate for skin with average permeability. This is shown by a journal article by employees at Alza, which assessed the degree to which the rate-controlling membrane controls fentanyl delivery across the skin.[FN9] This study stated that fentanyl delivery is “50% rate controlled,” which means that half of the resistance to fentanyl delivery is due to the patch and half is due to the barrier imposed by normal skin. This means that fentanyl delivery across the skin without the rate-controlling membrane of the patch would encounter only half the resistance, which would result in a rate of delivery that is twice as high. This value is different from the up to 5.6-fold increase indicated by my calculation above, but is nonetheless confirmatory of the conclusion that loss of the rate-controlling membrane significantly increases the fentanyl delivery rate.

    FN9. SK Gupta, Journal of Pain and Symptom Management (1992) 7:S17-S26

Although the average person may experience this increase of fentanyl delivery rate, by definition half of all people will experience a larger increase. This is shown in a clinical study carried out at McGill University, in which a fentanyl patch lacking a rate-controlling membrane was applied to the skin of 15 patients.[FN10] Three of these patients (i.e., 20%) had significant fentanyl toxicity in the form of respiratory depression and feeling dizzy, which made the researchers stop the study early. These effects were explained by the large variability in rates of fentanyl absorption among the patients, which resulted in a 20-fold variability in fentanyl concentration in the blood. The variability when using a non-defective Duragesic patch is just 4-fold, due to the presence of its rate-controlling membrane.[FN11] This clearly shows the dangers of transdermal delivery without a rate-controlling membrane, especially for those people with high skin permeability to fentanyl.

    FN10. P Fiset et al. Anesthesiology (1995) 83:459-469

    FN11. JR Varvel et al., Anesthesiology (1989) 70:928-934

The variability of skin permeability to fentanyl was specifically studied at the University of Michigan using human cadaver skin.[FN12] This study found that the skin's permeability to fentanyl varied 5.8-fold among the skin samples analyzed. A separate study was carried out at the University of Southern Denmark, which showed a 5.0-fold variability in skin permeability to fentanyl and an 8.5-fold variability in cumulative fentanyl absorbed across human cadaver skin.[FN13] To explain reported deaths caused by fentanyl patches, the authors stated that “this heterogeneity may render some individuals more susceptible than others due to unexpected high dermal absorption rate.” This study also measured the variability of skin absorption of fentanyl from Duragesic patches with intact rate-controlling membranes and found this variability to be just 0.5-fold. This again reinforces the idea that loss of the rate-controlling properties of an intact patch is dangerous, due to the large variability of skin permeability to fentanyl among different patients.

    FN12. SD Roy and GL Flynn, Pharmaceutical Research (1990) 7:842-847

    FN13. RH Larsen et al., Pharmacology & toxicology (2003) 93:244-248

l.b. Effect of skin damage on skin permeability to fentanyl

We have so far considered the effects of inherent variability in skin permeability, which suggests that loss of the rate-controlling properties of an intact patch would put a patient at greater risk that would be especially dangerous if that patient's skin permeability were greater than average. We should also consider the possibility that a patient's skin could have been damaged, which would increase skin permeability even more. In another study from the University of Michigan, the role of stratum corneum as a barrier to dermal absorption of fentanyl was examined.[FN14] This study determined that human cadaver skin lacking stratum corneum has a 67-fold higher permeability to fentanyl compared to intact skin. In this way, even a small defect in the thin layer of stratum corneum on the skin's surface can make a large difference in skin permeability. Small defects in stratum corneum can easily occur during normal daily life due to minor scratches.

    FN14. SD Roy et al., Journal of Pharmaceutical Sciences (1994) 83:1723-1728

A final consideration takes into account that fentanyl formulation leaking from a Duragesic patch can spread over an area larger than the patch, which will increase fentanyl delivery in direct proportion to skin area. Skin area is known to be a strong determinant of transdermal drug delivery, which is widely recognized in the transdermal delivery literature as well as based on well-established theories of diffusion. Indeed, the dosing method used in the Duragesic patch product line is based on this fact, such that delivery over the range of 12 to 100 mcg/hr is achieved by varying patch size from 5 to 40 cm2. Upon leaking out of the patch, fentanyl gel can spread over skin areas such as 100 cm2 (i.e., approximately 4 x 4 inches) to 1000 cm2 (i.e., approximately 12 x 12 inches) which would increase fentanyl absorption by 3.3 to 33-fold relative to delivery from a 30 cm2 patch. These skin areas of gel spreading are given here as examples and are not based on experiments or other information about the actual areas over which fentanyl gel from a leaking Duragesic patch spreads.

In summary, increasing fentanyl delivery by a few-fold relative to delivery from an intact Duragesic patch can produce lethal effects. The lethal threshold may be achieved by simply removing the rate-control membrane, which is expected to increase fentanyl delivery by 2-fold to 5.5-fold. The lethal threshold may also be achieved by an above-average skin permeability due to inherent skin variability, which has been shown to vary by 5.0-fold to 5.8-fold. The lethal threshold may also be achieved by an increase in skin permeability due to a nick or abrasion to the stratum corneum, which has been shown to increase 67-fold upon removal of stratum corneum. The lethal threshold may also be achieved by increasing skin area contacting the leaking gel, because fentanyl delivery rate increases with contact area. Finally, the lethal threshold can be achieved by a combination of these effects of (i) loss of the rate-controlling membrane, (ii) increased skin permeability and (iii) increased skin area.
II.c. Effect of solvent evaporation on transdermal fentanyl delivery

Another factor that has not been considered in this analysis is that if the fentanyl formulation were to leak out of a Duragesic patch, the ethanol-based solvent could evaporate. Below I will describe results from a number of studies from different research groups that clearly show that fentanyl delivery into the bloodstream continues even after its solvent has evaporated and that bioavailability of fentanyl delivered in this way can be high.

The first study was performed at London Hospital Medical College in collaboration with Janssen Pharmaceutica to measure transdermal delivery of fentanyl dried onto the skin.[FN15] In this study, an aqueous fentanyl solution was applied to the skin and allowed to rapidly dry by evaporation. Fentanyl absorption into the skin was then assessed over a 96-hour period. The study concluded that up to 53% of the fentanyl applied to the skin surface was absorbed within 24 hours and continued to be excreted in the urine for at least 96 hours. This study in human subjects clearly shows that fentanyl is absorbed into the bloodstream for many hours after its formulation has dried onto the skin surface.

    FN15. PS Sebel et al., European Journal of Clinical Pharmacology (1987) 32:529-531

Additional studies performed by Acrux Ltd. have reached similar conclusions. The core technology of this company involves ethanol-based formulations of drugs that are sprayed onto the skin. Although the solvent evaporates rapidly, drug delivery has been shown to continue for up to 24 hours or longer. In a poster presented at an international conference, fentanyl was applied and dried onto human cadaver skin using an ethanol-based solvent, which is similar to the ethanol gel found in Duragesic.[FN16] Despite rapid evaporation of the solvent, steady fentanyl delivery was measured over a 24-hour period. As further evidence that this phenomenon commonly occurs, this company also found that transdermal delivery continued over a 24-hour period after applying and evaporating ethanol-based formulations of other drugs, such as buspirone and granisetron applied to human cadaver skin and buspirone and testosterone applied to the skin of living human subjects.[FN17] Similar findings have additionally been presented in a number of patents issued to the company.[FN18] [FN19] It is worth noting that although this company is developing chemical enhancers that are added to the formulations to increase drug delivery, the data cited above is all for their “control” formulations that lack any chemical enhancers beyond the ethanol-based solvent.

    FN16. http://www.acrux.com.au/IRM/Company/ShowPage.aspx? CPID=1051&PageName=Modulation of the rate of transdermal drug delivery for treatment of pain with fentanyl', ‘Modulation of the rate of transdermal drug delivery for treatment of pain with fentanyl’, accessed September 23,2006

    FN17. http://www.acrux.com.au/IRM/Company/ShowPage.aspx? CPID=1054&PageName=Perspectives In Percutaneous Penetration',‘Perspectives In Percutaneous Penetration’, accessed September 23, 2006

    FN18. US Patent 6,818,226 B2

    FN19. US Patent 6,916,486 B2

These observations may be explained by a study from Monash University that measured the dermal absorption of testosterone after application of an ethanol formulation containing testosterone to swine skin.[FN20] Although this experiment did not use fentanyl, the similar physicochemical properties of testosterone and fentanyl suggest that these two drugs are delivered into the skin by a similar mechanism. After just one minute, 51% of the testosterone applied to the skin surface entered into the skin. This occurred because this drug is a small, oil-soluble molecule that can readily enter into the lipids of stratum corneum. In addition, during solvent evaporation, the fentanyl concentration can dramatically increase due to supersaturation of the drug in the ethanol formulation. Transdermal delivery from supersaturated solutions of a number of different drugs has been shown to increase the delivery rate by as much as 19 times.[FN21][FN22][FN23] The company, Acrux Ltd., refers to this as forming a “patchless patch,” such that the skin's stratum corneum rapidly becomes the drug reservoir for subsequent drug release from the stratum corneum into the deeper layers of skin and ultimately into the bloodstream.[FN24] This phenomenon is expected to occur not just for testosterone, but generally for small, oil-soluble drugs, such as fentanyl.

    FN20. TM Morgan et al., Journal of Pharmaceutical Sciences (1998) 87:1213-1218

    FN21. AC Williams, Transdermal and Topical Drug Delivery, Pharmaceutical Press, London (2003), pp. 114-122.

    FN22. NA Megrab et al., Journal of Controlled Release (1995) 36:277-294.

    FN23. MF Coldman et al., Journal of Pharmaceutical Sciences (1969) 58:1098-1102.

    FN24. http://www.acrux.com.au/IRM/content/technology/default frame.html, accessed September 23, 2006

    FN25. Physicians' Desk Reference, Thompson PDR, Montvale, NJ, 2006

This analysis can be further extended to a quantitative level. As a point of reference, Duragesic-75 administers 75 mcg/h of fentanyl across a skin area of 30 cm2. This corresponds to a transdermal flux of 2.5 mcg/cm2/hr. As discussed above in the Janssen study, fentanyl in an aqueous formulation applied and dried onto the skin of human subjects delivered up to 26.5 mcg over a 24-hour period.[FN26] Because the skin area was not given, we can assume that the 4 mcL of fluid formed a hemispherical geometry on the skin surface, which would correspond to a skin contact area of 0.05 cm2. This corresponds to a flux of 22 mcg/cm2/hr, which is almost 10 times greater than the flux of Duragesic-75. This clearly shows that fentanyl that has rapidly dried onto the skin surface can deliver fentanyl into the blood stream at a rate much higher than that of an intact Duragesic patch, which could cause a surge in fentanyl blood levels.

    FN26. PS Sebel et al., European Journal of Clinical Pharmacology (1987) 32:529-531

As a second quantitative comparison, the Acrux study states that fentanyl in an ethanol formulation sprayed onto human cadaver skin administered 12 mcg of fentanyl over a 24-hour period.[FN27] Based on the skin contact area of 1 cm2, this corresponds to an average flux of 0.5 mcg/cm2/h. Even though an extremely small dose of fentanyl was applied to the skin in this study (i.e., 1/625th of the fentanyl present in a Duragesic-75 patch), the flux from this evaporating formulation was 1/5th of the flux from an intact Duragesic-75. It should be noted that this flux calculated for the fentanyl spray is for the “control” formulation reported in the study without the company's proprietary enhancer, which most appropriately mimics exposure to fentanyl gel from the Duragesic patch.

    FN27. US Patent 6,916,486 B2

Finally, Alza carried out a study to measure the delivery of fentanyl from Duragesic gel dried onto the skin of rabbits.[FN28] The study concluded that fentanyl absorption from a fentanyl formulation dried onto the skin does not deliver large doses of fentanyl due to evaporation of the solvent. This finding as at odds with the collection of studies carried out by Janssen and Acrux using human skin and in human subjects. It is worth noting that the Janssen and Acrux studies were carried out by highly reputable research groups and were published in leading peer-reviewed journals, presented at international conferences, and disclosed in issued patents. The Alza rabbit study has been prepared only as an internal, confidential report and never published externally or subjected to peer review. The copy of this rabbit study made available to me contained only a rough outline of the experimental methods and the only information provided about the experimental results was a single graph with no text to accompany it. This makes the study's conclusions extremely difficult to assess, given the incomplete preparation of the document.

    FN28. Alza Corporation, Request for Testing BIO-01-B022-3400, RFT 12, March 19, 2004.

This rabbit study may have found anomalous results because it used rabbit skin. Rabbit skin is generally considered a poor model for human skin permeability and is rarely used in the peer-reviewed literature[FN29][FN30]. Moreover, in the poorly written study protocol, it appears that the thick fur of the rabbit was shaved a few days prior to the fentanyl delivery study. Some residual fur and the partial re-growth of fur after a few days could explain the low rates of fentanyl delivery observed in this study, since the gel would be caught up in the fur and make poor contact with the skin. In contrast, the intact Duragesic patch could overcome this limitation, since it was presumably pressed firmly against the skin and held in place by the adhesive that covers its surface. Thus, the Alza rabbit study is flawed not only in its incomplete presentation, but also in its fundamental experimental design. For all of the reasons discussed above, I would place confidence in the highly reputable results of the published Janssen and Acrux studies and question the significance of the Alza rabbit study.

    FN29. RC Wester and HI Maibach in Topical Drug Bioavailability, Bioequivalence and Penetration (VP Shah and HI Maibach, eds.), Plenum Press, New York (1993) pp. 333-349.

    FN30. R Panchagnula et al., Methods and Findings in Experimental and Clinical Pharmacology (1997) 19:335-341. A final issue that is particularly relevant to the DiCosolo case concerns the time scale over which fentanyl absorption occurs. As discussed above, fentanyl can be rapidly absorbed into the skin from an evaporating formulation. This absorption can occur on a time scale as rapid as minutes or possibly faster. In this case, the fentanyl is being rapidly absorbed into the stratum corneum layer of the skin. The fentanyl then forms a depot within the stratum corneum and is released from the stratum corneum much more slowly into the bloodstream. This release process can occur on a time scale of hours. This interpretation is consistent not only with the studies described above, but also with the pharmacokinetic studies presented in the Duragesic package insert.

In the case of Janice DiCosolo, the patch she was wearing at the time of death could have been responsible for her fentanyl overdose. She applied that patch approximately 6 hours before she was last seen alive and approximately 12 hours before she was found dead. This time scale is consistent with the expected time course of fentanyl absorption into the bloodstream from a depot formed in the stratum corneum from a leaky patch. The patch she was wearing prior to that patch could also have been responsible for her fentanyl overdose. Decedent's husband observed that this prior patch was poorly adhered to the skin and a slick residue was left behind on the skin after patch removal. This would be consistent with a leaking patch. The fact that decedent died 6 – 12 hours after removing that prior patch is also consistent with the expected time course of fentanyl absorption into the bloodstream from a depot formed in the stratum corneum from a leaky patch.

In summary, there is compelling evidence in the literature from a number of sources to show that fentanyl applied and then dried on the skin can deliver fentanyl into the bloodstream for many hours with fluxes that are similar or even greater than those achieved by an intact Duragesic patch.
III. Existence of a safer alternative patch design prior to the decedent's death

As discussed above, leakage of fentanyl gel from a Duragesic patch can be fatal and therefore careful measures should be taken to avoid it. Recognizing this, Alza's manufacturing procedures for the Duragesic patch are designed specifically to prevent producing leaky patches. However, even under normal operation, Alza recognizes that some leaky patches will be produced. Alza's quality control documentation indicates that it is acceptable for critical defects in Duragesic patches to occur at a rate of up to 1 critically defective patch per 10,000 patches manufactured.[FN31] Leaky patches are classified as “critical defects.” Given that an estimated 104 million Duragesic patches were sold in 2004,[FN32] this means that manufacture of approximately 10,400 critically defective Duragesic patches could have occurred that year under normal, acceptable operations.

    FN31. Alza Corporation, In-Process Packaging Inspections (SOP No. 890-0201), February 10, 2004.

    FN32. Estimate based on annual Duragesic sales of $2,083 million in 2004 (http://www.jnj.com/investor/documents/reports/2006annual.pdf) and a representative average price of $20 per patch (http:// www.walgreens.com/library/finddrug/druginfo.jsp?particularDrug=Duragesic, accessed Aug 6, 2007 and http://www.gb42.com/ynotfentany112point5.html, accessed Aug 6, 2007)

In an effort to prevent these critically defective patches from being given to patients, Alza's manufacturing procedures include a quality control inspection system. However, it has been documented that this inspection process was flawed. Leaky Duragesic patches were not detected by the inspectors and were then sold and used by patients. The FDA became involved and oversaw a recall in 2004 of a number of lots of Duragesic patches.[FN33] The first lot that was recalled, i.e., lot no. 0327192, was shown to have fold-over and other defects sufficiently dangerous to warrant recalling all patches in that lot. The defects in that lot were estimated to result in the release of thousands of defective patches to patients. Both the patch worn by Janice DiCosolo at the time of her death and the prior patch worn by her were from that defective, recalled lot.[FN34] Unfortunately, Alza did not recall this lot until after Janice DiCosolo's death.

    FN33. Janssen Pharmaceutica Products, Urgent Class 1 Drug Recall Notification Patient Level, February 17, 2004.

    FN34. Affidavit of John DiCosolo, November 17,2007.

In contrast to the reservoir design of the Duragesic patch, there exists an alternative matrix patch design that does not pose the risk of leakage and is therefore safer. In the Duragesic reservoir design, a fentanyl gel is encased within the patch, covered on one side with an impermeable backing, which is sealed to the rate-controlling membrane on the other side. If the backing is not correctly sealed to the rate-controlling membrane, as was the case during the 2004 recall, then fentanyl gel can leak out with catastrophic effects. Other types of defects can similarly lead to dangerous gel leakage. In the alternative matrix patch design, there is no fentanyl gel reservoir. Instead, the fentanyl is incorporated within the solid patch itself. There is no gel to leak out. Therefore, there is no risk of fentanyl leakage like there is in the Duragesic reservoir design. The advantage of the matrix design is widely recognized and specifically highlighted, for example, in Alza's Citizen Petition to the FDA, which states “matrix products do afford some advantages over reservoir products in terms of cosmetics, adhesion and eliminating the possibility of gel leakage (emphasis added).”[FN35]

    FN35. Susan P. Rinne, Alza Corporation, Citizen's Petition to the FDA, November 12,2004.

This alternative matrix design has been in existence for many years and is now in widespread use for fentanyl delivery in the United States and around the world. Many other drugs are also delivered using matrix-design patches and have been available in the United States since the 1980's. As early as March 2000, Alza was in communication with the FDA to discuss advanced stages of gaining approval of their matrix-design fentanyl patch in the United States.[FN36] Another meeting with the FDA to discuss Alza's matrix fentanyl patch was planned for April 2001 and Alza planned to submit a New Drug Application (NDA) for the matrix fentanyl patch to the FDA during the second quarter of 2002.[FN37] Later, Alza's efforts to gain FDA approval of a matrix fentanyl patch in the United States were abandoned. However, Alza continued to pursue approval of their matrix fentanyl patch in the rest of the world. By January 2004, Alza's matrix fentanyl patch was approved for sale in Germany[FN38] and was being marketed by May 2004.[FN39] In November 2003, another company (Mylan Technologies) received approval from the FDA for their matrix-design fentanyl patch,[FN40] although its marketing was delayed due in part to regulatory interferce by Alza In August 2006, a third company (Lavipharm Laboratories) received approval from the FDA for their matrix-design fentanyl patch.[FN41] This shows that well before the decedent' death, matrix-design fentanyl patches were being intensively developed by Alza and other companies and this technology was a feasible commercial alternative.

    FN36. Letter to Dr. Cynthia McCormick (FDA) from Kim Baumer (Alza), March 14, 2000.

    FN37. Internal Alza emails dated April 2001, DUR0121267 – DUR0121269.

    FN38. Information concerning the recently posted United States Food and Drug Administrations's Public Health Advisory regarding the safe use of transdermal fentanyl patches, DUR0097295 – DUR0097296.

    FN39. R. Freynhagen, et al., Journal of Pain and Symptom Management (2005) 30:289-297.

    FN40. http://www.fda.gov/ogd/approvals/ap1103.htm, accessed August 1, 2007.

    FN41. http://www.fda.gov/ogd/approvals/ap0806.htm, accessed August 4, 2007.

Comparing the Duragesic reservoir design and the alternative matrix design shows that under normal operation, the reservoir and matrix designs are equally good. This issue was explicitly addressed in a study by Alza comparing their reservoir and matrix fentanyl patchens.[FN42] This study concluded that the two patch designs were “bioequivalent” in all respects, which is the technical term for establishing that two drug systems perform equivalently. The FDA, as well as European and other regulatory agencies, have agreed with this conclusion, as shown by their approval of the various matrix-design fentanyl patches. Although these two patch designs are equivalent under normal circumstances the Duragesic reservoir design is clearly inferior under the unintended circumstance of fentanyl gel leakage.

    FN42. Sathyan et al. Current Medical Research and Opinion (2005) 21:1961-1968.

In summary, matrix-design fentanyl patches are superior to Duragesic reservoir-design patches, because matrix-design patches cannot leak. Matrix-design fentanyl patches were commercially feasible well before the time of the decedent's death. Therefore, if the decedent had been provided with a matrix-design fentanyl patch instead of the Duragesic reservoir-design patch, this would have eliminated the risk of fentanyl leakage from a defective patch and the associated risk of drug overdose.
IV. Additional Information

Over the past four years, I have not testified or been deposed in any cases other than previous cases involving the possible malfunction of Duragesic patches. I have served as an expert witness in other ways, as detailed in my curriculum vitae.

I declare the above to be true to the best of my knowledge under penalty of perjury.

September 18, 2007

1:30 p.m.

600 Peachtree Street

Suite 5200

Atlanta, Georgia

Kendra B. James, B-2194


For Plaintiffs:


Ferguson, Stein, Chambers, Gresham & Sumter, P.A.

741 Kenilworth Avenue, Suite 300

P.O. Box 36486

Charlotte, North Carolina 28236





Van Law Firm, P.A.

9930 Monroe Road, Suite 103

Matthews, North Carolina 28105





For Defendants:



1150 Huntington Building

925 Euclid Avenue

Cleveland, Ohio 44115





Also Present:

Mr. Harris Bitman, Videographer



(Whereupon, Exhibits 1 though 7 have beeen attached to the original transcript.)

THE VIDEOGRAPHER: We are now on the record. Today's date is September 18, 2007, and the time is 1:37 p.m.

This will be the deposition of Dr. Mark Prausnitz taken in the matter of Jason Burchill, et al., versus Johnson & Johnson, et al. Will the court reporter please swear in the witness.

MARK PRAUSNITZ, having been first duly sworn, was examined and testified as follows:



Q. Good after — good afternoon, Dr. Prausnitz. My name is Tariq Naeem. We've met before.

A. Yes.

Q. We're here today, as the court reporter mentioned, in the deposition — I'm sorry — deposition in the case of Jason Burchill versus Johnson & Johnson. I'm — I represent the defendants Johnson & Johnson, Alza Corporation and Janssen Pharmaceutical.

You've had your deposition taken before —

A. Yes.

Q. — I know that. Just a few quick ground rules to — to — to go

[Note: Pages 6-18 missing in original document]

A. It — it can. There — there certainly is literature to show that if the skin is made wet with water, it is more permeable. In the fentanyl patch I don't know to what extent the water plays that role.

Q. So, then, primarily, under these circumstances with the Duragesic patch, its primary role is for the solubility of fentanyl?

A. I think that's right.

Q. And the role of the hydrox — the — the hydroxyethyl cellulose, which I think is —

A. I — I'd refer to that as — as a gelling agent.

Q. Gelling agent.

A. So that — that's its function, is —

Q. Okay.

A. — is to form a gel rather than a — a fluid running liquid.

Q. And we're talking about, I guess, two 25-microgram patches.

Do you know the total amount of the volume of the contents of a 25-microgram patch?

A. The — the — the total fentanyl content or —

Q. No, just —

A. — the total —

Q. — the — the — the entire —

A. Size of the reservoir?

Q. How much gel — how — yeah, how much gel total is — is in a Duragesic 25-microgram patch?

A. I — I don't have that number with me.

Q. Then as to the constituents, then, how about the — what — what's the volume of fentanyl in a 25-microgram patch?

A. I believe I have that number in my report. No, I'm afraid I don't. I — I — I know where I could find that material, that — that number, but it's not given here. The concentration is what I have, but I don't have the volume or the absolute amount.

Q. Okay. So when you say “concentration,” what do you mean by —

A. So concentration is the — the — the amount, the mass of fentanyl divided by the — the volume of the reservoir. So —

Q. Okay. So —

A. The two numbers you have divided by each other.

Q. Okay.

A. The — the two numbers you asked for divided by each other.

Q. All right. Then, so what is the concentration of fentanyl in a Duragesic patch — I'm sorry — a 25-microgram Duragesic patch?

[Note: Pages 21-59 missing in original document]

Q. Would you agree with me that the fentanyl solution in Duragesic gel is saturated?

A. It is saturated certainly at the time a Duragesic delivery begins. Toward the end of the experiment, as the fentanyl becomes depleted and it goes across the skin, there could be a point where it is no longer saturated.

Q. This may be simplistic. I apologize.

But does the presence of Duragesic in suspension give you an indication that the fentanyl in the solution is saturated?

A. If you have fentanyl that is in suspension, then probably the — the fentanyl in solution is saturated.

Q. And just so we are on the same page, what does — can you define for me “saturated”; what does it mean when a solution is saturated?

A. It means that you have dissolved as much of the fentanyl in the liquid as can be dissolved.

Q. And what — what does the term “supersaturated” mean?

A. Supersaturated refers to dissolving more fentanyl in solution than would normally be allowed. But sometimes you can get more to dissolve in, usually for some period of time. But if you wait long enough — and long enough could be seconds and long enough could be ??